The Bench · N° 41

What a timegrapher actually measures

Rate and beat error come from the timing of the escapement sounds. Amplitude is calculated from that timing, so the lift-angle setting matters.

Hen & Mills Editorial · 4 min read · 17 September 2026

A timegrapher does not look inside a movement and measure every part of its performance. It listens to the escapement, identifies a repeating pattern and turns that pattern into a few useful numbers.

The familiar display usually shows rate, beat error and amplitude. Two are closely tied to event timing. The third is a calculation that depends on a setting entered by the operator.

That distinction matters when you interpret the result.

What the microphone hears

Each oscillation produces a short sequence of sounds as the pallet fork and escape wheel interact. The machine detects those impulses through a contact microphone attached to the watch or movement holder.

It then plots their timing as dots or lines. A steady, well-adjusted movement often produces clean parallel traces. Noise, interference or an irregular escapement can produce scattered or broken patterns.

The screen is a diagnostic view of the escapement's rhythm, not a complete health certificate for the watch.

Rate

Rate is the estimated daily gain or loss, usually shown in seconds per day. The machine compares the measured beat frequency with the frequency the movement is meant to have.

If the beats arrive slightly early, the displayed rate is fast. If they arrive late, it is slow.

This reading is useful, but one position and a few seconds of measurement cannot describe a full day on the wrist. Mechanical watches often run differently dial up, crown down and in other positions.

Beat error

Beat error measures the difference in timing between the two halves of the balance's motion. In an ideal beat, the impulse events are evenly spaced around the balance's neutral position.

The result is normally shown in milliseconds. A small reading means the two halves are close to symmetrical. A larger reading can point to an adjustment issue.

Beat error is not the same as rate. A watch can show a reasonable rate while still having excessive beat error, and correcting beat error can alter the rate.

Amplitude

Amplitude describes how far the balance swings from its resting position, expressed in degrees. It helps a watchmaker judge how much energy is reaching the oscillator.

Unlike rate and beat error, the machine does not directly watch the balance wheel turn through that angle. It calculates amplitude from the timing of the sounds and the lift angle entered for the movement.

That makes the setting critical. Enter the wrong lift angle and the displayed amplitude will also be wrong.

What lift angle means

Lift angle is the portion of the balance's arc during which the escapement is delivering impulse. Its value is determined by the movement's geometry.

Many movements use a value near 50 degrees, but there is no universal setting. Similar-looking calibres can specify different figures.

The best source is the movement manufacturer's technical information. If the lift angle is unknown, the amplitude reading should be treated as an estimate and labelled accordingly.

Why amplitude changes

A fully wound movement usually has more available torque than one near the end of its power reserve. Position, lubrication, dirt, balance condition and escapement adjustment can also affect the swing.

A low number can suggest that energy is being lost, but it does not identify the cause by itself. An unusually high reading can also need investigation, particularly if the escapement is knocking.

Context matters: movement design, state of wind, position and lift-angle setting all belong beside the number.

Reading the trace

A clean diagonal line generally indicates a consistent rate offset. Two separated parallel lines represent beat error. Scatter or repeated disturbances can indicate noise, an unstable beat or a mechanical issue.

The pattern is often more informative than a single headline number. A trained watchmaker looks at how the trace behaves, whether it settles and how it changes between positions.

Phone applications can provide a rough rate check, but their microphones and signal processing do not always isolate the escapement well enough for dependable amplitude or beat-error work.

Why one position is not enough

Gravity changes how pivots and the balance behave in different orientations. A movement may look excellent dial up but run differently crown down.

A useful check records several positions, with the movement adequately wound and allowed to settle after each change. The chosen positions should match the purpose of the test, whether that is workshop regulation or understanding normal wrist wear.

Averaging positions can be helpful, but it should not hide a single position that behaves abnormally.

What a healthy-looking result cannot prove

A timegrapher cannot confirm water resistance, magnetism resistance, power-reserve duration or the condition of every wheel and bearing. It also cannot show how the watch behaves through an entire wind-down cycle from one brief reading.

A recently serviced movement can produce poor numbers if it was measured incorrectly. A worn movement can briefly show respectable numbers in a favourable position.

The machine supports diagnosis. It does not replace inspection or judgement.

How to make a useful comparison

Use the same state of wind, lift angle, position and settling time for each reading. Record the beat rate and measurement duration as well.

If you are comparing before and after service, compare like with like. If you are monitoring a watch at home, look for a pattern over time rather than reacting to a single result.

The most useful question is not whether one number looks impressive. It is whether the readings are credible, repeatable and appropriate for that movement in those conditions.